资源分类
》 资源 》论文》Adaptive Attention Span in Transformers

Adaptive Attention Span in Transformers

2019-09-22 | |  74 |   41 |   0 0 0
Abstract We propose a novel self-attention mechanism that can learn its optimal attention span. This allows us to extend significantly the maximum context size used in Transformer, while maintaining control over their memory footprint and computational time. We show the effectiveness of our approach on the task of character level language modeling, where we achieve state-of-the-art performances on text8 and enwiki8 by using a maximum context of 8k characters.

上一篇：A Surprisingly Robust Trick for the Winograd Schema Challenge

下一篇：Analyzing Multi-Head Self-Attention: Specialized Heads Do the Heavy Lifting, the Rest Can Be Pruned

用户评价
全部评价

热门资源

  • The Variational S...

    Unlike traditional images which do not offer in...

  • Learning to Predi...

    Much of model-based reinforcement learning invo...

  • Stratified Strate...

    In this paper we introduce Stratified Strategy ...

  • A Mathematical Mo...

    Direct democracy, where each voter casts one vo...

  • Rating-Boosted La...

    The performance of a recommendation system reli...